The project develops an innovative approach and a highly versatile, fully automated apparatus (new technology and new methods) to test for the effects of genetic and pharmacological manipulations on the mechanisms of memory. The new system makes possible large scale psychophysical screening for manipulations that cause systematic discrepancies between what is stored to and what is later retrieved from memory (between what goes in and what comes Out). Because these discrepancies are proportional to the quantities remembered, they are called "scalar errors." Previous research indicates that rat, pigeon, and human subjects demonstrate scalar errors in their memory of temporal intervals. The magnitude and sign of scalar errors are affected by neurobiological disease processes (Parkinson's); can be altered pharmacologically; and differ between subjects - differences that may be of genetic origin. Current memory screening approaches typically measure the strength, probability, frequency, and time to acquisition of a conditioned response - measures that conflate memory with performance factors. These new measures may prove superior in that they quantify distortions that the underlying memory mechanism introduces into the contents of memory. Individual mice inhabit a fully automated, self-cleaning plus maze. The mouse nests in the center of the plus and forages in darkness, visiting each arm 300-800 times per night. A flexible and scriptable computer program determines all cues and environmental factors, including the positions and times at which food is delivered in each arm. The subject's position in the arms is detected 50 times per second by infrared beam curtains with a spatial resolution of 3/4 inches. This level of spatial and temporal precision, combined with the numerous visits per night, makes possible efficient and accurate determination of distributions regarding latencies and locations. Exactly where and when a mouse shows up for food depends on its particular memory for where and when it has previously been fed. Through automation, the project effectively examines outbred (wild-type) strains of mice to measure scalar error (and scalar variability) in their memories. By determining the range of these memory parameters, and by correlating them across the spatial and temporal domains, the project develops a body of psychophysical data on which significant future genetic and pharmacological screening research may be built.